1 Introduction
Scent from a buried source moves through the soil, ground cover, scent boundary layer, and into the air where it can be detected (Figure 1). The availability of scent in the air above a buried source depends on the properties of the soil and scent molecules and their interactions; on processes that occur in the soil, ground cover, and scent boundary layer; and on the weather. Scent plume movement in air is influenced by terrain and vegetation, and weather influences scent movement through all media between the source and the dog. Scent movement through the soil is constrained because it can move only in the pore spaces between soil particles while in the gas and liquid phases. Scent transport also depends on the rate of movement and on whether the VOCs are in the gas phase or the water phase (i.e. how they are distributed or partitioned between phases). In dry soil, VOCs are adsorbed on the soil particle surfaces as a kind of residue and are not free to move. Our primary interest is in upward movement of scent, but downward movement can be of interest especially in the presence of flowing underground water.
Information on the movement of scent in soils is available from studies of buried explosives and decomposition VOCs, but there does not appear to be any for drugs and other buried sources. Fortunately, there is a substantial literature on the movement of gases, water, fertilizers, pesticides, and other chemicals in agricultural soils and on the movement of hazardous contaminants in soils. This information can be used to develop insights on scent movement in soils that are useful in training and deploying SDs used for detecting and finding explosives, drugs, cadavers, and other buried sources.
The use of SDs to detect buried sources is a difficult task that tests the team’s ability to detect and locate the faint scent of a source that may be wrapped or in a container, covered by different soil types, ground surface covers, vegetation, terrain, and in variable weather conditions. The scent molecules available at the soil surface are known for some explosives (Jenkins et al. 2000). Scent molecules for cadavers (Vass et al. 2004, 2008, 2012) have been cataloged, but the specific types used by the dogs to detect cadavers are not generally known. While some information is available, more is needed to determine the quantitative effects of soil, ground cover, and surface environmental conditions (primarily weather, vegetation, and terrain) that influent the success of SDs in finding buried sources. This chapter is an introductory discussion of scent transport in soils and attempts to show how this information can be exploited under certain conditions to increase PODs for buried sources. The results are not completely satisfactory because they are based on anecdotal information, sparse data, limited theory, and untested hypotheses.
Searching for a clandestine grave
Figure 1 Scent from a buried source must pass through soil, ground cover, scent boundary layer, air, vegetation, and over terrain to reach the dog’s nose. Weather influences scent movement through all these media.
1.1 Soil and VOC Properties
Soil consists of inorganic particles from weathered rock, organic products from the fora and fauna that exist in or on the soil, and materials transported by water and wind (Marshall and Holmes 1979). Soil particles (grains) vary in size from coarse-grained gravels and sands to fine-grained silts and very fine-grained clays. Soil particles are in contact with each other and have three-dimensional interconnected pore spaces that create channels through the soil (Figure 2). These pore spaces contain soil gases (air and water vapor) and soil water or ice. Scent (VOCs) is present as gases, dissolved in water, insoluble liquids (from wet sources), and as source micro-particulates. Soil water also exists as thin water films on soil particle surfaces. Gases and liquids move through the channels formed by the pore spaces in the soil and water films move over the surfaces of the soil particles. Scent can move through soil as gases in the soil pores, solutes dissolved in the water films and pore water, insoluble liquids, and as solids in the form of source micro-particulates much smaller than the pore spaces (Figure 2). Scent in all its forms can also enter the groundwater and be transported by it.
Physical properties of interest for scent movement in soils include soil type, porosity, saturation, water content, temperature, and organic content. Soil types include organic and inorganic (gravel, sand, silt, clay) and mixtures of them. Porosity is the fraction of the total soil volume that consists of pore space. Silts and clays have greater porosity (e.g. clays, about 0.6) than sands and gravels (e.g. sands, about 0.4), but sands and gravels have much larger interconnected pore spaces which make them much more permeable to the flow of gases and liquids (Table 1). The degree of saturation is the fraction of pore space filled with water. Soil is 100% saturated when the pore space is filled with water. A dry soil may have just 1% of its pore space filled with water. Dry soils are common in arid climates but are usually limited to the surface soil in temperate climates. Water content is the fraction of the total soil volume or weight that is water. In some areas, especially lowlands, a water table may exist. Soil is unsaturated above the water table and saturated below it.
Figure 2 A conceptual two-dimensional model of soil containing air and water. Scent can move through soil pores by gas phase diffusion and advection or transported when dissolved in pore water and water films that move through the pores and over the surfaces of the soil particles.
Soil temperatures influence all the chemical and physical properties of soils and scent. Temperature influences volatilization rates (availability of scent) largely through its effect on the vapor pressure of VOCs. It also influences the movement of scent molecules in both the gas phase and when dissolved in the water. Organic content is the fraction of the total soil volume or weight that consists of organic material and typically ranges from a few per cent to about 80%. Soils with large amounts of organic material adsorb more scent than mineral soils which leaves less available for transfer to the air. This makes it somewhat more difficult for dogs to detect sources buried in soils with large organic contents. Clay content and soil pH influence volatilization through adsorption effects that reduce the solution concentration of dissolved VOCs and VOC concentrations in the gas phase.
Table 1 Relative Values of Porosity and Flow Compared for Gravel, Sand, Silt, and Clay
The chemical properties of scent molecules that influence scent availability and movement include vapor pressure, phase partition coefficients, diffusion coefficients, solubility, and degradability. Vapor pressure is a measure of the volatility of a chemical and can vary significantly with commonly encountered changes in soil temperature. VOCs with high vapor pressure produce more scent than those with low vapor pressure.
The concentrations of VOCs in soil air, dissolved in soil water and attached to soil particle surfaces, depend on the phase partition coefficients for two phases in contact and at equilibrium. For example, the partition coefficient for a VOC in soil air in contact with soil water containing the dissolved VOC is the concentration of the VOC in the air divided by the concentration in the water. These coefficients are determined primarily by the chemical compound, soil type, water content, and temperature. Phase partition coefficients determine whether scent transport to the soil surface will be in the gas or water phase. For example, the primary chemical components of TNT include TNT, DNT, and dinitrobenzene (DNB). The phase partition coefficients for these chemicals in air in contact with water are small which indicates only a small fraction of a percent exists in the air with about 10% in the water phase and the rest adsorbed on soil particle surfaces that are not free to move (Phelan and Webb 2002). Since the water contains almost all the mobile chemicals, water phase transport dominates scent movement in the soil for these chemicals except at low water contents.
The phase partition coefficients for decomposition chemicals vary considerably. Since it is not known what VOCs are used by dogs in detecting buried cadavers, it is not possible to thoroughly evaluate decomposition scent movement in soil and scent concentrations in air over buried cadavers.
Diffusion coefficients are useful for determining the amounts of VOCs that diffuse in the air and water phases. Diffusion coefficients for VOCs in air are roughly 10,000 times greater than for those dissolved in the water (Phelan and Webb 2002). However, as for explosives, the concentrations of VOCs in each phase that are available to move must be considered and these depend on the phase partition coefficients.
Water solubility is the amount of VOC that can dissolve in water. There are large differences in the solubility of explosives, drugs, and decomposition chemicals. For example, at room temperature, TNT is only slightly soluble (0.13 g/L) while AN is highly soluble (150 g/100 ml). Degradation processes in soils cause source VOCs to change form, chemical properties, and can result in their elimination from the soils (Phelan and Webb 2002). Soil mineral reactions, biological transformations, and plant root uptake are examples of these processes. Degradation rates depend on soil water content, soil type, and temperature and are characterized by their half-life (time required for the VOC concentration to decrease by half). Some VOCs degrade quickly with lifetimes in soils measured in days. For example, the relatively low concentrations of 2,4,6-TNT in soils compared with its other chemicals appear to be attributable to a very short half-life for this compound (about a day) compared to the half-life for 2,4-DNT (about 26 days) under similar conditions, which indicates that 2,4-DNT is more likely to survive passage to the soil surface.
Water contents more than 1% result in fast degradation rates and the effects of clay or organic contents in soils is to reduce the half-life of VOCs. Subfreezing soil temperatures (<32°F) and low water contents, much <1%, limit degradability. Increasing soil temperatures from freezing to about 72°F can decrease the half-life for 2,4-DNT by a factor of 3 or more in a wet soil.
There are significant differences in the chemical properties of VOCs associated with buried sources, especially vapor pressure, phase partition coefficients, diffusion coefficients, and degradation rates. These differences indicate that the concentration of each VOC in the gas and water phases will be different, each VOC will move with different diffusive velocities in soil air and water, and the lifetime of each VOC in the soil will be different. This implies there will be a spectrum of VOCs above buried sources ranging from those that are absent to those with enough concentration to be detectable by dogs.
1.2 Buried Sources
Burying a source involves digging a hole, placing the source, fling the hole, disposing of excess soil, and camouflaging the surface disturbance. FBI data (Hofman et al. 2009) shows that clandestine burial sites were located a short distance from infrequently traveled roads or pathways, approximately 10 f away from the closest large tree and were typically surrounded by bushes or heavy foliage. Typical burials varied in depth from 1.5 to 2.5 f and the corpse was usually clothed or wrapped in plastic and faced down. Difficulties associated with digging also suggest that the grave will not usually be placed where large rocks or hard dry clay are found. If the source is heavy such as an adult body, it will usually be brought to a point near the grave by a vehicle and carried not much more than about 100 yds (Rebmann et al. 2000) unless there are multiple people involved or the corpse is light.
The soil placed in the hole will have significantly different properties, especially structure, density, and water content compared to the surrounding undisturbed soil. This indicates that scent movement from the source to the surface in the disturbed fill soil will be modified compared to that in the surrounding undisturbed soil. Scent movement to the surface will likely be enhanced in the fill soil because it is difficult to compact the fill soil to the same degree as undisturbed soil. The characteristics of the scent distribution surrounding a buried source depend on the physical and chemical properties of the soil and VOCs, whether the soil and source are dry or wet and whether the source VOCs are soluble or insoluble. Buried explosives would normally be dry and recently buried cadavers would be wet.
If the soil and source are dry or the source is insoluble, dissolved fluids will not exist. If the soil is wet with water soluble VOCs, soil pore water and soil water films will contain dissolved VOCs. If the soil contains insoluble liquids, their fate depends on whether they are lighter or heavier than water and how they interact with soil particles. Some decomposition fluids are relatively insoluble in water and often lighter. When a cadaver is buried in contact with the soil, interaction of the decomposition fluids with the soil fines can produce a mucus sheath around the cadaver consisting of the fluids and fines soil fraction (Janaway 1997). Little is known about the details of this process.
Figure 3 is a conceptual model of the possible scent distribution associated with buried human remains based primarily on studies of explosives and hazardous chemicals in soil. The scent can move through the soil except for that adsorbed on soil particle surfaces, on organics, and that removed from the flow by processes discussed above.
Figure 3 A conceptual model of a grave showing the possible scent distribution on a slope. Decomposition fluids occur under the body. The mucous sheath and micro-particulates are not shown.
The ground surface and the groundwater table can be level or sloped. If both are level, scent transported to the surface over buried sources can accumulate during favorable conditions and create a surface distribution of scent that is expected to be slightly larger than the source and centered over the source.
On sloped ground, the surface scent distribution would extend downslope from the source as shown in Figure 3. Aitkenhead-Peterson et al. (2011) have observed signifcant downslope movement of decomposition chemicals from cadavers on the ground surface. Figure 4 shows the soil surface concentrations of 2,4-DNT downslope from land mines buried under a gently sloping ground surface (Hewitt et al. 2001).
Since 2,4-DNT is the primary chemical in TNT that dogs use for detecting the explosive, the results shown in Figure 4 indicate that the spatial distribution of this chemical in the surface soils is a “scent print.” The TMA-5 mine is roughly 7 × 8 in and produces a scent print roughly 2 × 2 f or about 3 times the size of the mine on this slope. The highest values of scent do not occur at the mine but about 6 to 8 in downslope from the mine. The PMA1A is about 3 × 4 in and produces a scent print about 1.7 f wide, extending 2.3 f downslope or about 6 to 7 times the size of the mine on this slope. The highest values of scent do not occur at the mine but a few inches downslope from the mine. Reasons for the differences in the results for the mines are unknown but likely associated with mine characteristics, soil properties, surface slope, and other effects.
Figure 4 Spatial distribution of 2,4-DNT (ng/g) in surface soils downslope from buried land mines, at Ft. Leonard Wood, Missouri. A TMA-5 antitank mine (left) and PMA1A antipersonnel mine (right). (From Phelan and Webb 2002. Data from Hewitt et al. 2001. See also Osterkamp 2020.)
The importance of the observed high surface concentrations of 2,4-DNT associated with mines buried in dry soils is that the concentration available for detection is orders of magnitude higher (thousands of times) on the surface soil particles than in the air. On dry soils, this suggests dogs may be inhaling soil particles suspended by the action of air jets to detect mines and/ or desorbing VOCs on the soil particles with their humid exhalations. While inhalation of surface soil particles over landmines and desorption of VOCs by their humid breath may be alternate mechanisms that dogs use for detection, it is not necessary because of their robust capabilities for trace vapor detection. Nevertheless, these scenting processes may be used during favorable environmental conditions.
It is thought that the size of the surface scent print for a mine that is buried on level ground should be about twice the lateral dimensions of the mine. On slopes, surface runoff can create a scent print for some distance downslope of the source (Figures 4 and 5). Sarginson et al. (2012) found that scent was transported downslope from buried mines in small eroded surface drainage channels during severe rainfall events resulting in numerous false alerts by EDs in the channels. This is further evidence of a scent print associated with mines. If erosion channels are present on slopes, scent from buried sources may be concentrated in them and these features should be searched carefully. SDs may alert downslope of the source so that an excavation carried out at the point of an alert may not fnd the source. It is useful for the dog to be trained to find the position of the strongest scent which is most likely close to the source (Figure 4). Strategies to narrow the source location are to work the dog from different directions, especially downslope toward the initial alert, use multiple dogs, return under more favorable or different conditions and, if the source is not hazardous, vent the soil by probing upslope of the initial alert.
If a buried source is in unsaturated soil above a water table, the vapor plume, dissolved soil water plume, insoluble liquids, and micro-particulates may eventually penetrate to the water table as shown in Figures 3 and 5. Insoluble liquids and micro-particulates lighter than water would foat on the water table and the dissolved soil water plume would mix with the water. Buoyant and dissolved materials would be transported with groundwater flow. Non-buoyant liquids and micro-particulates would sink below the water table while being carried along by groundwater flow.
If a source is buried below the water table (not common) or the water table rises above the source after burial, buoyant insoluble liquids and micro-particulates would be expected to rise to the water table and be transported downstream by groundwater flow. Non-buoyant insoluble liquids and micro-particulates would move downward by gravity and downstream in groundwater flow. In either case of a source above or below the water table, if there is groundwater flow, scent can be transported underground for long distances.
Figure 5 Scent print transported downslope of a buried source by surface run off. Underground water flow can transport scent to downslope seeps or springs that are often found at the toe of slopes and along the shores and banks of lakes and rivers.
Vass (2012) reported decomposition VOCs at a distance of 1/2 mile from human remains. Scent can be brought to the surface in seeps and springs that occur on slopes often at the toe of the slope (Figure 5) and along the banks of lakes and rivers, usually near the water level. These seeps, springs, and lake and river banks in a search area should be carefully checked by SD teams for the presence of scent.
2 Scent Movement in Soils
2.1 Processes
Scent movement to the surface from sources buried at shallow depths (typically <2 f) is complex, involving soil and source properties, a host of processes associated with physical, chemical, biological components, and weather effects. The complexity and level of scientific background required to address these processes and lack of information on buried sources, except for explosives and some contaminants, make it difficult to develop an understanding of scent movement associated with buried sources. The following is an attempt to present scent transport processes in terms that are more accessible and familiar to SD handlers. For the most part, observable information (soil, vegetation, terrain, and weather) is used to evaluate site conditions, potential search times, and search methods. It is a limited qualitative approach that does not address all the processes involved. However, it may allow handlers to develop a better understanding of scent transport in some of the conditions and settings where they work.
For more information, Jenkins et al. (2000) summarized the results of laboratory and feld research from an extensive study of buried land mines at Ft. Leonard Wood, Missouri. Phelan and Webb (2002) provided a comprehensive summary of the behavior of TNT, DNT, and DNB from mines in soil which includes laboratory and field research, modeling of scent transport, availability of scent at the soil surface, and the use of EDs. A detailed feld study of the efects of environmental and other variables on the detection of mines by EDs in Afghanistan was conducted by Sargisson et al. (2012). Rivett et al. (2011) reviewed the movement of VOCs, primarily contaminants in unsaturated soils. Some of these contaminants are associated with decomposition and other sources.
Transport processes for VOCs in unsaturated soils from buried sources are broadly classified as diffusion and advection in the gas and water phases as shown in Figure 6. Difusion is the movement of VOCs at the molecular level in all directions (radially) away from the source. It is ofen the dominant mode of scent transport in the gas phase. Advection refers to the bulk movement (currents) of air and water containing scent.
Evapotranspiration refers to evaporation of water at the soil surface and transpiration of water from vegetation. These processes dry the soil at and near the surface and produce a gradient in water content that pulls (wicks) water and dissolved scent upward. Infiltration of precipitation by gravity moves water and dissolved scent downward. Transport of micro-particulates could occur in the water phase and downward by gravity but there is no information on this topic. If the source is wet, such as a recently buried cadaver, the possible presence and transport of insoluble liquids would have to be considered.
Figure 6 A conceptual model of VOC vertical transport processes in unsaturated soil. Vertical arrows represent processes that result in vertical fow upward or downward. Horizontal arrows represent processes that redistribute the phases or remove VOCs from the fow. (Modified from Phelan and Webb 2002.)
Horizontal double ended arrows (Figure 6) represent reversible partitioning between gas, water, and solid phases because of volatilization/ dissolution and adsorption/desorption. There are several degradation processes that remove VOCs from the fow (single ended arrows in Figure 6) by transformation and loss by chemical reactions, microbial biodegradation, and plant root uptake. The amount of scent flowing to the surface can be significantly reduced by these processes, especially if the half-life of the VOC is short compared to the time required for it to reach the surface.
On exiting the soil surface, the scent passes through any ground cover and a surface boundary layer into the atmosphere where conditions can vary greatly between day and night and seasonally.
2.2 Gradient Transport in Unsaturated Soil
VOCs in the gas phase and those dissolved in pore water and water films on soil particle surfaces can move in the air and water by diffusion and be carried by advection in moving air and water. Computer-assisted modeling is required to predict the complexities of scent movement. A limited understanding can be obtained by considering the effects of changes in quantities that are observable under field conditions. In this simplified approach, scent movement is a result of gradients (differences) in temperature (T), moisture (M), pressure (P), and chemical concentration (C) of VOCs (Marshall and Holmes 1979). Like heat, scent flow is in the direction from high values to low values. The movement is influenced by the magnitude of the gradients, properties of the VOCs, soil properties, and the effects of external conditions, primarily weather. The effects of gravity on micro-particulates and insoluble liquids such as insoluble decomposition fluids depend on whether they are lighter or heavier than water and have been discussed above.
Chemical properties are unique to the VOC, vary significantly between VOCs, and depend on temperature and water content, among other factors. These considerations make it problematic to extrapolate information for one VOC to others. Soil properties are strongly dependent on soil types and water content. Variations in weather can influence scent transport at shallow depths in the soil by causing changes in temperature, water content, pressure, and chemical concentrations over all time scales.
Gradient movement can occur both upward and downward. Our primary interest is in upward movement which requires that surface values be less than those at depth. These gradients are responsible for moving air and water containing VOCs to the surface where they move from the near-surface soil into ground cover and a boundary layer in the air above.
Temperature gradients transport VOCs in air and water from warmer to cooler soil so that cooler surface temperatures compared to the source temperatures result in scent transport out of the soil which creates favorable search conditions. These conditions can be caused by nighttime cooling of the ground surface, short term weather changes, and seasonal effects. If the source is not too deep (about a foot or two depending on soil properties), nighttime surface cooling can cause temperature gradients later at night and during early morning, with source temperatures warmer than the ground surface temperatures. These temperature gradients initially cause water vapor, gas phase VOCs, and water films containing VOCs to move toward the surface and create favorable conditions for scent flow out of the surface in the late night and early morning hours before the sun heats the surface. SD handlers have noted improved detection of buried sources at these times.
Soil temperatures can lead to competing effects in scent transport. For example, an increase in temperature causes an increase in volatilization rates but concurrent soil drying with adsorption of VOCs on soil particle surfaces may not result in an increase in the availability of scent. Degradation rates are greater at warmer temperatures, but at colder soil temperatures volatility is lower and less scent is available. Freezing conditions may halt VOC degradation. Degradation by-products may be detectable by SDs but there is little information on this topic.
Dry soils typically adsorb large amounts of VOCs on soil particle surfaces and leave a residue that reduces available scent levels. As the water content increases, water replaces the VOCs adsorbed on the soil particle surfaces and releases them for possible transport to the surface and atmosphere. As the soil becomes wetter, gas pore space is reduced, gas phase transport decreases, and water phase transport increases.
Laboratory experiments have shown that increasing the humidity of dry soils can release huge quantities of VOCs adsorbed on the soil particle surfaces (Petersen et al. 1996). Surface flux changes of about 3 orders of magnitude (103 = 1000 times) for DNT have also been observed during wetting and drying events (Phelan et al. 2001). This indicates that favorable times to search for sources buried in dry soils are late night and early morning when air humidity is high, when dew is present, or after a light rain. Informal experiments have shown that misting a dry soil surface can substantially improve scenting conditions for explosives and suggest that this may also be possible for other buried sources provided that the soil/air partition coefficients are large for the VOCs associated with these sources.
Pressure gradients can transport scent out of the soil surface, provided the pressure at the surface is less than the pressure at the source and the scent is in the gas phase. These pressure gradients can be caused by reductions in barometric pressure, rising tides, wind, and other factors. Consequently, search conditions are more favorable when the barometer is falling or low, tides are rising in tidal zones, and when the wind is strong and variable.
Chemical gradients transport scent in the gas phase and solutes in the water phase from regions of higher to lower scent concentrations (i.e. away from the source in all directions including upward flow toward the surface and downward below it).
In the absence of precipitation, evapotranspiration tends to dry the uppermost soil. The resulting gradient in water content draws water vapor and water in films containing dissolved VOCs to the soil surface where evaporation continues. Evaporation is a distillation process that leaves the VOCs behind where they may volatilize directly into the air as scent or be adsorbed on soil particle surfaces to be released later under favorable conditions. Evaporation may be largely responsible for the surface scent print observed over mines.
VOC half-lives decrease rapidly with increases in soil water content from dry conditions and as soil temperatures increase above freezing. Some hydrocarbons associated with decomposition have half-lives less than a year. If the half-life of source VOCs is much less than the time required for them to move from the source to the soil surface, the amount that reaches the surface may be much less than that available near the source and some VOCs will never reach the surface.
The presence of the soil surface interrupts transport of VOCs upward and can result in accumulation of water and VOCs there. This can modify the gradients in water content and chemical concentrations and even cause them to reverse locally and result in a local backflow downward. Condensation or evaporation at the surface may also occur and modify the gradients and make it impossible to thoroughly evaluate scent transport in the field.
Scent movement can be restricted artificially. Landfills that are compacted daily and then covered by specially designed soil caps that do not allow flow of gases including scent through them are an example. Searches for bodies in these landfills is an almost hopeless task unless the cap is disrupted, the garbage removed in layers and made accessible to the dogs, and the dogs are allowed access to underlying material. This type of search is hazardous to the dogs.
3 Soil–Air Interface
3.1 Boundary Layers
The boundary layers of interest for buried sources are associated with surface roughness elements such as pebbles and large rocks and vegetation consisting of leaves, grass, weeds, and bushes. These roughness elements produce turbulent boundary layers somewhat greater than their height above the ground surface. There are several types of boundary layers over buried sources including those for water vapor and scent. For smooth, fat, and level bare soil with no wind, and surface temperature less than air temperatures (inversion present), the scent boundary layer thickness for buried explosives is thought to be less than an inch (Phelan and Webb 2002). Boundary layer characteristics are influenced by the ground cover and weather (primarily wind speed and radiation) and possibly by the actions of the dog exhaling (Figure 2.3) and movement that create temporary local air currents just above the surface. Boundary layer thicknesses decrease with increasing wind speed. Solar radiation, when present, warms the ground surface which warms the air in contact with it. Colder air above with warmer air at the surface is an unstable condition that results in small scale thermal convection that mixes scent from the ground surface, with the air above possibly moving it upward out of reach of dogs. The magnitude of this effect depends primarily on the temperature difference between the soil surface and the air.
On bare soil, the effects of air stability and wind on scent from a buried source start to become important as the scent exits the soil and boundary layer and enters the atmosphere (Figure 7). For stable conditions with soil surface temperatures cooler than the air and no wind (typical of conditions at night, when the surface is in shade and cloudy days), an inversion may form and scent would be expected to pool over the source in a thin inversion layer. If a small depression is present the thickness of the scent boundary layer may be much greater than that for a fat surface, probably as thick as the depth of the depression. When ground cover is present it would also be thicker, probably approaching the thickness of the ground cover. If the ground cover is dense, it may be useful to wait for convective instability or wind that would help move the scent upward so the dog could detect it. With sufficient wind, the scent layer would be disrupted, and a scent plume would form downwind where it would be more easily detected.
Figure 7 Scent movement over bare ground above a buried source showing the effects of solar radiation, air stability, and wind on the scent plume. The surface is cooler than air temperatures for stable conditions and warmer for unstable conditions.
Stable calm conditions require the dog to search with its nose close to the ground surface and to be almost directly over the source to detect the scent. These search requirements are similar when a depression exists over an older grave since the depression traps scent in it during the night under relatively calm conditions. The scent remains in the depression until it is removed by wind or convective turbulence. These conditions make it desirable to select search dogs for buried sources that show a tendency for nose to ground searching and to use training methods that produce dogs proficient in this type of searching. Hound breeds and some dogs of other breeds do this naturally.
For unstable conditions with soil surface temperatures warmer than the air and no wind (typical of conditions when the surface is in sunlight with your shadow <1½ times your height), convective instability would form a vertical plume over the source. With wind speed much greater than the rise velocity of air in the plume (typically about 1 f/sec), the vertical plume would be bent toward the horizontal where it would be more easily detected.
For unstable and calm conditions, the dog’s nose could be elevated but it would have to pass through or close to the vertical plume over the source to detect it. The above conditions indicate that search lanes would have to be narrow, perhaps only a few yards for large sources and even smaller for small sources, to obtain a high POD. Stable and unstable conditions with wind that produces a scent plume at the dog’s level allow SDs to detect the plume some distance downwind with its head elevated using much wider search lanes.
3.2 Ground Covers
Ground cover effects are associated primarily with low lying vegetation in contact with the soil such as live organics (plants, moss, grass, weeds), dead organics (leaf litter, dead plants), and the effects of a transient snow cover. Bare ground is exposed to atmospheric conditions while ground cover mitigates their effects on the underlying soil by shading the soil, insulating it, and reducing the amount of solar radiation that reaches the surface. This results in lower surface temperatures and decreased evaporation from the surface. Ground cover can absorb water, increase surface wetness, and influence water content in the subsurface. It influences scent behavior when scent exits the soil surface and passes through the ground cover into the air.
CDs often show a change of behavior when sniffing grasses, weeds, shrubs, and especially tree trunks near a buried body. There are two hypotheses to explain this behavior. One is that VOCs from the buried source are taken in by plant roots, carried to the stem and leaves, and volatilize into the air where they can be detected. This is a complex process that requires nutrient VOCs from the source to contact the roots, be absorbed, travel through the plant to the leaves, and volatilize into the air. The second is that VOCs present in the air above a buried source collect on nearby plant surfaces where they can be dislodged by air jets from the dog’s nose or volatilized and be detected by dogs sniffing these surfaces. While it is known that chemicals can be absorbed through the roots and enter plants (Carter et al. 2014), the second hypothesis is simpler but there does not appear to be any data confirming either hypothesis and one or both may be true.
Information on the characteristics of a scent plume moving from the soil into a surface cover is limited. Sargisson et al. (2012) showed that detection success decreased with increasing vegetative cover near a mine. If the vegetation allowed the surface to dry and was sufficiently sparse so that dogs could get their noses close to the ground, then a mine search could still be done. Spiky plants and plant smells near the mine had no significant effects on detection success for sparse plants.
Informal experiments with a cover of grass suggested that the boundary layer thickness was about the same as the thickness of the grass cover. For article searches in tall grass in Florida, it has been hypothesized that sunlight was needed to produce convective (turbulent) transport out of the grass (Mesloh and James-Mesloh 2006). Convection would facilitate scent movement upward, in and above grass for the dogs to detect scent without having to place their noses deep in the grass. This may also be true for other types of dense vegetation (taller grass, weeds, small shrubs). Scenting conditions with little or no wind and vegetation higher than the dog’s head are very difficult. Experience with hunting dogs suggests that a dog must almost step on a bird to find it under these conditions. There does not appear to be any information on the boundary layer associated with leaves or needles on a forest floor.
Snow covers can be classified for our purposes as wet or dry, windblown or not, and compacted snow. Wet snow covers are associated with snow that falls under warm conditions, with daily warming, or warming weather events that cause internal and surface melting. In some regions, melting during daylight hours and refreezing at night is common. This can produce impermeable ice lenses in the snow that retard or prevent flow of scent to the snow surface.
Dry snow covers are associated with cold climates and/or high elevations where subfreezing temperatures are common and continuous. It is a low-density snow that favors scent movement. Dry snow is such a good insulator that a thickness of 2 f can cause ground surface temperatures to rise to a few degrees below freezing with subzero air temperatures. This creates a large temperature gradient that favors scent transport out of the snow. Windy conditions cause snow to drift. The action of wind on snow is to fragment snow crystals into tiny particles that bond together to form a hard, dense surface layer of snow (wind slab) that retards the movement of scent.
Compacted snow includes avalanched snow, ploughed snow berms along roads, and snow dumps that result from clearing travel routes and other areas in northern cities. Snow dumps can be dangerous to search because chemicals, needles, broken glass, and other hazardous materials are collected with the snow. Compacted snow severely retards scent transport. However, dogs have demonstrated the ability to find live people and cadavers in avalanched snow within minutes of being buried. It is proposed herein that this ability is a result of the settling that occurs when a flowing avalanche consisting of snow and a large volume of air stops moving. On settling, most of the air would be forced out of the snow to the surface carrying the scent of a buried victim with it. The expired air would leave a trapped scent plume in the snow above the victim and scent on the surface of the snow that dogs could detect. This process would not apply to slab type avalanches.
Generally, subfreezing air temperatures cause the soil water in pores to freeze from the surface downward. Except for dry soils, freezing partially or completely blocks the movement of scent. Moss, vegetation, and dry snow are good thermal insulators, and act like a blanket that tends to keep the soil surface warmer than the air temperature. Consequently, soil under a layer of moss, vegetation, or snow may be unfrozen even when air temperatures are below freezing so that searches for buried sources in these conditions can sometimes be successful. Probing the soil is the only sure way to determine if the soil is frozen or not.
When there is little or no snow cover with subfreezing air temperatures, fine grained frozen soils tend to contract and crack. Contraction cracking can allow scent to escape directly to the atmosphere provided that the cracks penetrate to the buried source. Cracks deeper than 3 ft have also been observed to form in agricultural soils under hot and dry conditions, and when they occur in a search area, they should be checked for scent. If source scent is detected above a crack, the source may be directly below the alert or some distance from it because of possible lateral movement of scent in the crack.
4 Searching for Buried Sources
This section examines the effects of soil, vegetation, and weather on searches for buried sources and compares conditions that would result in lower or higher PODs. The effort is hampered for at least two reasons. First, except for some explosives and drugs, it is not generally known what chemicals are used by the dogs to detect specifc sources. Since the chemicals are unknown, their physical and chemical properties that influence scent movement cannot be determined. For example, phase partition coefficients determine the relative quantities of the chemicals in each phase (gas or liquid) and adsorbed on soil particles. If the chemicals are unknown, the coefficients are unknown so that scent movement and adsorption effects in soils cannot be evaluated. Second, the influence of different types of ground cover, vegetation, and small-scale thermal turbulence on scent movement and plume behavior are not well known which makes it difficult to evaluate scent movement from the soil surface to the dog’s nose. Consequently, the following comparisons on the effects of soil, weather, and vegetation on detector dog searches may change as more information becomes available. The information used in these comparisons is partly anecdotal (primarily from K9 handlers), a result of laboratory and field research, and from modeling studies.
Jenkins et al. (2000) have shown that the chemicals associated with TNT in mines are distributed unevenly around the mine, decrease rapidly in concentration upward, and some may not reach the ground surface. The gradient processes described above are responsible for upward movement of the chemicals through the soil. Degradation processes cause decreasing concentrations upward that can prevent some chemicals from reaching the surface. These gradient and degradation processes are controlled by the properties of the soil and scent and by the weather which collectively determine the type, amount, and rate of chemical movement to the ground surface. Consequently, it is necessary to consider these processes, properties, and weather effects to understand how source chemicals move from the source to the ground surface.
The results of Jenkins et al. (2000), Hewitt et al. (2001), and Sargisson et al. (2012) indicate that DNT from buried mines is concentrated in the surface soil, forming a scent print. Te sources of VOCs evolving into the air above buried sources are the VOCs adsorbed on the surface soil particles (i.e. in the scent print), VOCs in the gas phase that transit to the soil surface from below, and VOCs produced by volatilization of soil water at the soil surface. The relative importance of site and environmental conditions on these processes that produce detectable scent for SDs are not well known which makes it difficult to evaluate the best times to perform searches.
It has been noted (The Dog’s Nose and Scent) that the scenting methods used by SDs to detect sources involve direct sniffing of source VOCs and actively using exhalation air jets to extract VOCs from surfaces and particulates. Exhalation air jets on surfaces that are dry and dusty or have loose particulates on them disturb the surface materials so that dogs can inhale them and extract adsorbed VOCs. It is not likely that this process would be efcient on moist surfaces. This implies that soil, weather efects, and other factors determine the amount of scent available for detection. These considerations may hold for other sources but there is little information available except for some hazardous chemicals that are also associated with decomposition.
4.1 Searching
Searches for buried sources are usually planned, except for the military, and a common strategy is to select a time when conditions would be optimal for dogs to detect and locate the source. In addition to the difficulties noted above, others exist partly because of a lack of quantitative information on what constitutes optimal conditions and partly because of uncertainty in predicted weather conditions at the time of the proposed search. For the most part, we can only defne search conditions qualitatively: those that have a lower probability of success and those that have a higher probability. We cannot state a reliable value for those probabilities.
Favorable times to detect a buried source are when working and scenting conditions are good and when the scent is likely to be available to detect. Working and scenting conditions are good when the dogs do not become physically or mentally stressed and poor when they do. Examples of poor conditions include high temperatures, dusty conditions, muddy ground, thick or thorny vegetation, steep terrain, and too much interference by the handler while the dog is searching. Te availability of scent above the surface of a buried source depends on its movement through the soil, ground cover, and as a plume in the atmosphere. If scent has accumulated in a scent print at the surface, its availability would depend on conditions favorable for volatilization at the surface (Cousins et al. 1999).
Scent from a shallow buried source spreads laterally in the soil as it rises to the surface where it can create a surface scent print somewhat larger than the source. Te size of the surface scent print is of interest to SD handlers since it helps to defne the width of the search lanes to be used. If the search lanes are too wide, POD for the source will be reduced. Lane widths for shallow buried sources appear to depend primarily on the horizontal dimensions of the source and weather (especially atmospheric stability and wind, see Figure 7). As the density of surface vegetation (grasses, weeds, prickly vegetation), depth of burial, and desired POD increases, lane widths should decrease. Wind makes it possible to use a greater lane width for a given set of conditions. The experience and reliability of the team can also be important in selecting lane widths.
Lane widths for SDs used in demining operations are set by the agency or country that oversees these operations (GICHD 2004). Te dogs are trained to work on long leads (about 9 to 12 yds) or on short leads. Long lead dogs work in lanes about 1.5 yds wide although the handler has some discretion in choosing the width. Tey are taught to go out from the handler to the end of the leash in the lane being searched, return in the previously searched lane, and move to the next unsearched lane. Short lead dogs work in lanes about 0.5 yds wide. Te handler walks in the previously searched lane while the dog on lead searches the next lane. The handlers attempt to work the dogs in lanes across the wind and may change the direction with changes in wind direction. Each lane is searched by 2 dogs. Other patterns include circular, semi-random, and free ranging.
Limited information on lane widths for buried bodies and other large sources appear to range from about 2 to 10 or more yds depending on conditions. Searches for bones, teeth, fred shell cases, and other small sources need smaller lane widths. In searches for teeth placed on the ground surface, 3 dogs searching on leash using a lane width just under 2 yds, and with cross gridding on a windy day (16 mph average speed) recovered a maximum of 78% of the teeth (Cablk and Sagebiel 2011).
Linear searches (along paths, trails, roads, edges of structures, or natural features like vegetation, fields, shorelines, etc.) are best done on the downwind side of the feature being searched. If there is a steep nearly vertical wall (formed by vegetation, terrain, or a structure), searches on the downwind side should be close along the wall. On the upwind side, they should be some distance away from the wall.
Sun shining on a wall, steep surface, or vertical vegetation (e.g. trees, Figure 3.1) heats the surface which heats the adjacent air and causes it to rise along the surface. Under calm conditions, this efect carries scent near the wall upward and out of reach of a dog. For windy conditions, sun and wind produce opposing air currents on the upwind side and reinforce air currents on the downwind side which makes it difficult to evaluate the resulting scent movement.
A method often used in area searches by CD handlers is to first do a free ranging search through the area. The team starts on the downwind side with a coarse grid and checks any likely places for a burial. If nothing is detected, the handler then uses a fine grid with lanes running crosswind.
Searches for some sources (e.g. explosives) can be extremely hazardous and require special methods.
4.2 Soil Conditions
Environmental conditions such as soil, weather, vegetation, and terrain are site specific and the availability of scent depends on the chemicals (VOCs) in the scent and how these interact with the local environmental conditions. This implies that results from field studies and modeling will be site specific and will depend on the type of source. Care must be taken when extrapolating results from one site to another and for results from one type of source to another when attempting to define optimum times and conditions for conducting searches. For example, computer modeling results and the conclusions of field studies for one type of source may not hold for different locations and climates or for VOCs associated with different sources. These characteristics, especially partitioning between phases, diffusion rates, evaporation, degradation, and loss in the soil, influence VOC movement in soils and the availability of scent in the air above the soil. Consequently, while the results shown in Tables 2 to 4 are based on general conditions, some may not hold for all sources in all places. Table 2 compares the efects of soil conditions, source size, and depth of burial on PODs for buried sources.
Gravity is always present and would move scent downward, especially during infltration. Lighter than air scent molecules in unsaturated soils and buoyant materials in saturated soils would move scent upward.
Large chemical concentration gradients exist near a source that can move scent rapidly by diffusion over short time periods (Rivett et al. 2011). Vertical movement as a result of chemical gradients will be upward above the source and downward below it. Above the source, movement by concentration gradients may be changed or temporarily reversed by temperature, moisture, and pressure gradients that move scent toward the source.
Weather conditions for shallow burials are important in determining whether the net movement will be upward or downward. Consider the gradients to result from the diference between the values at the buried source and at the ground surface. Maximum upward scent flow occurs when all the gradients near the surface cause upward flow (T, M, P at the surface are less than at the source) and minimum when all the gradients cause downward flow (T, M, P at the surface are greater than at the source).
These relatively simple conditions can become complex because of transient effects produced daily and by changing weather conditions that modify the near-surface gradients. During the day, solar radiation increases soil surface temperature and evaporation. At night, radiation to space decreases soil surface temperature and may result in condensation (dew). In addition, changing weather (barometer, wind, rain) can modify the near-surface gradients which can change near-surface scent concentrations. Scent movement is extremely difficult to determine when the separate gradients cause flow in different directions and computer-assisted modeling is required to evaluate it.
Table 2 Comparison of the Effects of Soil Conditions, Source Size, and Depth of Burial on PODs for Buried Sources
Lower PODs do not mean that the source cannot be detected but that detection is less likely than for higher PODs. Water content is that between the source and surface.
Increasing organic contents of soils increases the amount of scent adsorbed and leaves it less available for detection. However, the level of scent removed may not make it impossible to detect a source. Information on soil types can be obtained from agricultural soil maps and can be helpful in planning a search and interpreting the results. Fine-grained soils such as clays and silts are less permeable to scent fow than coarser soils like sand and gravel. Where soil types vary locally or are intermixed, this knowledge may be less useful. Ice in frozen soil can block the scent flow, as noted, unless the soil is dry.
Sargisson et al. (2012) have shown that detection success increases with increasing size of mines containing TNT and decreases with increasing depth of burial. The latter result may be attributed to degradation processes in the soil. Both results appear to hold for other buried sources.
4.3 Weather Conditions
Weather conditions that influence scent transport from buried sources to the atmosphere include wind, barometric pressure, air temperature, radiation, humidity, and precipitation including snow (Table 3). These conditions are important because they influence scent transport processes through their effects on soil conditions at depth, at the ground surface, in ground cover, and in the air above.
Table 3 Comparison of the Effects of Weather Conditions on PODs for Buried Sources
Lower PODs do not mean that the source cannot be detected but that detection is less likely than higher PODs. Note: h is not hours, it is the handler’s height.
4.3.1 Wind
Consider the effects of wind only. Calm (no wind) means there is no scent plume. Fortunately, this condition is rare. There is usually a slight wind drif with low wind speed and random direction that produces a meandering scent plume which makes it somewhat easier to fnd the source. Strong winds with swirling and gusting elements that distort, fragment, dilute, and redirect the scent plume make it difficult to fnd the source. Light winds (5 to 10 mph) produce a well-defned scent plume that make it easier to fnd and follow the plume to the source (France et al. 1997; Ruzicka and Conover 2011).
4.3.2 Pressure Gradients
Both diffusion and advection may be important in moving soil gases and scent to the surface. An increased outfow of soil gases (air, water vapor, scent) is expected with decreasing air pressure at the ground surface and increasing air pressure at the water table. Decreasing air pressure at the ground surface can result from a falling barometer and wind (Bernoulli efect). Increasing air pressure at the water table can result from a rising water table such as that caused by a rising tide. Whatever the cause, if there is little scent in the gas phase (as with TNT), additional scent fow due to advection would be small. Since the VOCs used by dogs to detect buried cadavers are unknown, partitioning between gas, water, and soil phases cannot be determined. This makes it impossible to evaluate the effects of air pressure in transporting scent to the soil surface over buried cadavers and many other sources.
Considering only wind and barometric pressure, if there is little or no wind, decreasing barometric pressure would dominate movement of scent in the gas phase to the surface. If there is constant barometric pressure with significant wind, wind effects would dominate, and if there is decreasing barometric pressure coupled with strong wind, pressure effects for bringing scent to the surface would be optimal but not necessarily significant.
It appears that wind dominates where significant wind occurs over agricultural soils. This may be due to tillage in these soils which makes them more permeable to the flow of soil gases. Soil permeability is also greater over buried sources because of the disturbed soil so that windy conditions may help bring scent to the surface over them. This also suggests that wind over coarse sands and gravels may be effective in bringing scent in the gas phase to the surface.
4.3.3 Air Temperatures
4.3.3.1 Warm Temperatures When scent concentrations are low and scenting conditions poor, panting may reduce the ability of SDs to detect a source. However, experience has shown that they can still fnd a buried source if they are close to it (within about a yard) and if the soil is moist (France et al. 1997; also see Figure 7).
Lasseter et al. (2003) conducted a study of CDs in Alabama during July and August when air temperatures ranged from 82°F to 93°F with humidity of 55% or more. Results showed that hot weather and high humidity always infuenced the performance of the dogs even with multiple breaks and water available. Handler experience suggests that at temperatures <65°F, dogs can do light work without thermal stress.
Handlers can cause emotional stress in SDs and reduce their performance by using too many commands, getting irritated with the dog, and failing to let the dog work.
4.3.3.2 Cold Temperatures and Snow Results from France et al. (1997) are that temperatures limit the ability of dogs to detect scent at a distance if the source is buried in soil. For sources buried in snow with air temperatures that allow little or no melting, the dogs had to be within a yard of the source to detect and fnd it. If there was signifcant melting, the dogs could locate the source from a greater distance. With sources buried in snow or in soil below snow, and air temperatures below freezing, the dogs may not be able to locate the source. While the details and data of this study are not fully known, it appears likely that snow thickness and the permeability of the snow to scent (determined by snow type and age, internal structure, previous melting and refreezing events, presence of ice layers) may have infuenced the results. Air temperatures below freezing are not always sufficient to freeze the ground since undisturbed snow is such a good insulator.
Aspens and Aspens (1998) conducted a study in Interior Alaska where snow is typically less dense (powder), not refrozen, has no ice lenses, and is more permeable. The results showed that dogs could locate a cadaver source on the ground surface buried in snow at temperatures of −25°F and at distances of 25 yds or more. Handler experience with SDs in Interior Alaska indicates that dogs have no difficulty performing scent work with sources buried in snow at temperatures less than −30°F.
The study of Komar (1999), which tested the ability of CDs to detect cadaver sources on the ground surface covered with litter, showed that temperatures warmer than −22°F and snow depths up to a foot had no effect on performance. Familiarity with sources and experience in these terrains and conditions did influence performance.
If the soil is moist and frozen from the surface downward, the presence of ice in the pore spaces would generally block scent from moving upward and make it difficult to impossible for dogs to detect a buried source. If the soil is dry and frozen, the pore spaces would not be blocked by ice and it may still be possible to detect a buried source.
4.3.4 Wind and Sun (Air Stability)
The effects of wind and atmospheric stability on scent plumes have been noted previously. France et al. (1997) found if there was no wind, dogs had difficulty detecting a source unless they were within about a yard of it. This result is expected when there is no wind for both stable and unstable conditions (Figure 7) and suggests that the scent print at the ground surface was about 1 to 2 yds across for the buried pigs used in the study. With enough wind (perhaps 3 to 5 mph) and a stable atmosphere, the layer of scent at the surface would be disrupted and form a scent plume downwind (Figure 7) which makes it more likely that the dogs would detect it. If the air is unstable with no wind, the scent plume would rise vertically and carry the plume out of reach of the dogs. Te dogs would have to pass through or close to the plume to detect it. Detection is easier if the wind speed is sufciently large to force the plume down near the level of the dog’s nose. Te study suggested that a wind speed of >5 mph is desirable although this result may be due to site conditions such as vegetation height and density that require a certain level of turbulence to move scent from the ground cover into a plume. When wind speed is high and the wind is gusting, extreme turbulence near the ground may distort, redirect, dilute, and fragment the scent plume and make it more difficult to detect it and to follow it to the source.
The length of a person’s shadow is a rough measure of the air stability as discussed in Scent and Wind. Stable air with no wind allows scent from a buried source to concentrate near the ground and stable air with wind produces a scent plume. Unstable air with no wind causes scent to rise vertically over the source and with wind produces a scent plume. Wind velocity much greater than the convective rise velocity can cause a vertical plume to be forced down to the level of the dog.
4.3.5 Humidity
Relative humidity depends strongly on temperature. Maximum relative humidity occurs about daylight when air temperatures are minimum and reaches a minimum about the time of maximum temperatures. Relative humidity near the ground surface can be dramatically diferent from that measured at chest height. If the relative humidity at chest height is 30% and the surface is 20°F warmer, the relative humidity just above the surface will be roughly half or about 15%. This makes it difficult to evaluate the impact of humidity and temperature measured at chest height on scenting conditions near the ground surface.
Studies on the effects of humidity on the ability of dogs to detect a buried source have produced complex and seemingly different conclusions. France et al. (1997) suggested that the humidity should be about 20% or higher for optimal scenting conditions in Colorado. Sargisson et al. (2012) conducted a detailed study of weather factors that infuenced the ability of EDs to detect and fnd buried mines in Afghanistan for 5 periods during a year. Searches were conducted during the morning daylight hours only. For the conditions unique to this region, it was concluded that wind speed, air temperature, and relative humidity had no overall significant effects on detection success although humidity was deemed the most important.
The results indicated that high humidity produced somewhat poorer detection in this arid environment, except in the early daylight hours, when high humidity and dew on the ground surface appeared to facilitate detection (possibly by releasing adsorbed VOCs from the dry surface soil). The cool surface also favors upward scent movement in the soil. Detection success typically decreased with decreasing humidity until about 9 a.m. and then increased with decreasing humidity until midday when testing was halted for the day. Te initial decrease in detection in early morning may have been caused by drying of the soil surface as it was warmed by the sun. A reduction in the availability of scent would have been caused by readsorption of VOCs and by reversal of the temperature gradient near the surface. Increased detection during later morning is thought to have been the result of the dogs actively desorbing surface scent molecules with their warm and humid breath. Another possibility is that the unstable air near the surface, as a result of solar heating, may have facilitated convective movement (thermal turbulence) upward of air and scent where it would be more accessible to the dogs.
The reasons for these apparently different results between the Colorado and Afghanistan studies have not been addressed. They could be due to the different physical and chemical properties of the VOCs associated with decomposition and explosives, presence of vegetation, small scale convective turbulence near the ground, scenting method used by the dogs, or the wicking effect associated with evaporation at the surface. It is known that humidity influences the availability of scent from explosives but there is a lack of quantitative information on the effects of humidity on other sources.
Humidity does not appear to have a significant effect on the scenting ability of the dogs over ranges encountered in the field (about 20 to 80%). Some handlers mist their dog’s noses when humidity is low to improve their scenting ability, an untested hypothesis.
4.3.6 Precipitation
High humidity, dew, and light rain on dry soils release the adsorbed VOCs on the surface soils into the atmosphere which increase the availability of scent there by orders of magnitude as noted. Heavy rain dilutes the concentrations of near-surface VOCs, moves them downward as it infiltrates the soil, decreases availability of scent, and lowers POD. Light powder snow is permeable to scent while dense, wet snow, especially that which has been refrozen or contains ice layers, has low permeability to scent.
Table 4 Comparison of the Effects of Vegetation on PODs for Buried Sources
Table 4 compares the effects of vegetation on POD for buried sources. Mesloh and James-Mesloh (2006) found that mid-mornings were the most productive time of day to search tall grass in Florida and hypothesized that the efect of dense vegetation on scent coming from articles on the soil surface was to allow scent to remain in the vegetation until wind and convective instability caused it to move upward and out of the cover. Typically, wind and instability do not develop until about mid-morning so this hypothesis suggests that in searches of areas that contain a significant vegetative ground cover, there should be an improvement in PODs during the morning.
If the ground cover is much higher than the dog’s head, it may not be possible for the dog to detect a plume above the cover. Sargisson et al. (2012) concluded that the spikiness of plants surrounding a mine had no signifcant efect on detection success. When vegetation is dense and contains prickly types such as cacti, thorns, or briars, dogs and handlers may be reluctant or not able to penetrate them and leave these areas unsearched.
5 Summary
Scent from a buried source moves through the soil, ground cover, scent boundary layer, and into the air as a plume where it can be detected by a dog (Figure 1). The availability of scent in the air above a buried source depends on the properties of the soil and scent molecules and their interactions; on processes that occur in the soil, ground cover, and scent boundary layer; and on weather which infuences scent movement through all media between the source and the dog’s nose. Scent movement through the soil is constrained because it can only move in the pore spaces between soil particles while in the gas and water phases. Scent transport also depends on the rate of movement and on whether the VOCs are in the gas phase or the water phase (how they are distributed or partitioned between phases). In dry soil, VOCs are adsorbed on the soil particle surfaces as a kind of residue and are not free to move.
Scent is present as gases, dissolved in soil water, as insoluble liquids (from wet sources), and as source particulates. Soil water exists in soil pores and as thin water flms on soil particle surfaces. Scent can move through soil as gases in the soil pores, as solutes dissolved in water flms and pore water, as insoluble liquids, and as solids in the form of source particulates much smaller than the pore spaces (Figure 2). Physical properties of soils that influence scent movement include soil type, water content, temperature, organic content, and others.
Soil temperatures influence the chemical properties of soils and scent volatilization rates (availability of scent) largely through their effect on the vapor pressure of VOCs. Temperatures also infuence the transport of VOCs in both the gas phase and when dissolved in the water. Soils with large amounts of organic material adsorb more scent than mineral soils which reduces the amount of scent available to move and makes it more difficult for dogs to detect sources buried in these soils.
The chemical properties of VOCs that influence scent availability and movement include vapor pressure, phase partition coefficients, diffusion coefficients, solubility, and degradability. Phase partition coefficients determine whether scent transport to the soil surface will be in the gas or water phase. For example, the primary components of TNT include TNT, DNT, and DNB. The phase partition coefficients for these chemicals in air in contact with water are small indicating that only a small fraction of a percent exists in the air with about 10% in the water phase and the rest adsorbed on soil particle surfaces that are not free to move. Since the water contains almost all the mobile chemicals, water phase transport dominates scent movement for TNT in dry soils.
Degradation processes in soils cause source VOCs to change form and chemical properties and can prevent them from reaching the soil surface. Soil mineral reactions, biological transformations, and plant root uptake are examples of these processes. Some VOCs degrade quickly with lifetime in soils measured in days. For example, the concentration of 2,4,6-TNT in soils is reduced by one half in about a day compared to about 26 days for 2,4-DNT indicating that 2,4-DNT is more likely to survive passage to the soil surface.
FBI data indicates that clandestine graves typically varied in depth from 1.5 to 2.5 f were located a short distance from infrequently traveled roads or pathways, about 10 f away from the closest large tree, surrounded by bushes or heavy foliage, and the corpse was usually clothed or wrapped in plastic and faced down. A conceptual model of the scent distribution associated with a grave is shown in Figure 3.
Since 2,4-DNT is the primary chemical in TNT that dogs use for detecting the explosive, Figure 4 indicates that the spatial distribution of this chemical in the surface soils is a surface “scent print” (Osterkamp 2020). If the ground surface over a buried source is level, the scent print is roughly circular, larger than the source and centered over it. On sloped ground, the scent print extends downslope from the source as shown in Figures 3 and 4. Figure 4 shows measured scent prints on gently sloping ground for antitank and antipersonnel land mines. The TMA-5 mine is roughly 7 × 8 in in size and produces a scent print roughly 2 × 2 f in size or about 3 times the size of the mine. The PMA1A is about 3 × 4 in and produces a scent print about 1.7 f wide and extends to 2.3 f downslope or about 6 to 7 times the size of the mine. This scent print from a buried mine is unevenly distributed, and, on slopes, the highest values occur several inches downslope from the mine with the scent print extending 2 f downslope within a year of burial. It is hypothesized herein that a similar scent print may exist for other buried sources.
If erosion channels are present on slopes, the surface scent from buried sources may be concentrated in them and these features should be searched carefully. These channels may cause EDs to alert significantly downslope of the source so that an excavation carried out at the point of an alert may not find the source. It is useful for the dog to be trained to find the position of the strongest scent which is most likely close to the source. Strategies to narrow the source location are to work the dog from different directions (especially downslope) toward the initial alert, use multiple dogs, return under more favorable or different conditions, and vent the soil by probing upslope of the initial alert. A dog should be present during the excavation of an alert site to provide direction for the digging effort.
Scent that penetrates downward to flowing underground water can be transported long distances underground (at least ½ mile for human remains). It can be brought to the surface in seeps and springs that often occur at the bottom of slopes and along the banks of lakes and rivers, usually near the water level (Figure 5). These seeps, springs, and lake and river banks in a search area should be checked by SD teams for the presence of scent.
Scent movement to the surface from sources buried at shallow depths is complex and involves soil and source properties and a host of processes associated with physical, chemical, biological, and weather effects (Figure 6). This requires computer modeling. A simplified approach is developed that uses handler observable information (soil, vegetation, terrain, and weather) to evaluate site conditions, potential search times, and search methods. In this approach, scent movement in unsaturated soil is a result of gradients (differences) in temperature (T), moisture content (M), pressure (P), and chemical concentrations (C) of VOCs. Like heat, scent flows from high values to low values and upward movement requires that surface values be less than those at depth (Table 2).
Temperature gradients transport VOCs in air and water from warmer to cooler soil so that cooler surface temperatures compared to the source temperatures result in scent transport upward, which creates favorable search conditions. These conditions can be caused by nighttime cooling of the ground surface, short term weather changes, and seasonal effects. Temperature conditions are usually favorable for scent flow to the surface in the early morning hours, during cold spells, and during seasonal cold weather.
Pressure gradients can transport scent out of the soil surface, provided the pressure at the surface is less than the pressure at the source and the scent is in the gas phase. Pressure conditions are more favorable when the barometer is falling, tides are rising in tidal zones, and the wind is strong and variable.
In the absence of precipitation, evapotranspiration tends to dry the uppermost soil. The resulting gradient in M draws water vapor and water in films containing dissolved VOCs to the soil surface where evaporation continues. VOCs on the surface may volatilize directly into the air as scent or be adsorbed on soil particle surfaces to be released later under favorable conditions. Informal experiments have shown that misting a dry soil surface can dramatically improve scenting conditions for explosives and suggests that this may also be possible for some other buried sources. This indicates that a good time to search for sources buried in dry soils is early morning when air humidity is high, when dew is present or after a light rain.
Chemical gradients transport scent in the gas phase and solutes in the water phase from regions of higher to lower scent concentrations (i.e. away from the source in all directions including upward flow toward the surface).
The effects of air stability and wind on scent from a buried source become important as the scent enters the surface boundary layer (Figure 7). For stable air conditions with soil surface temperatures cooler than the air and no wind (typical of conditions at night, when the surface is in shade and cloudy days), an inversion may form, and scent would be expected to pool over the source in a thin inversion layer. With these conditions, dogs need to search with their noses close to the ground. For unstable conditions, soil surface temperatures warmer than the air, and no wind, convective instability would form a vertical plume over the source. Dogs can search with their heads up. For both stable and unstable conditions with no wind, dogs must pass very close to or over the source to detect it. When wind is present, dogs can detect the source in the downwind plume.
Snow can be wet or dry, windblown or not, and compacted. Wet snow that refreezes at night usually has ice lenses that retard or prevent flow of scent to the snow surface. Dry snow favors scent movement to the surface. Windy conditions produce a dense surface layer (wind slab) that retards the movement of scent. Compacted snow from some types of avalanches severely retards scent transport.
Dogs can find live people and cadavers in avalanched snow within minutes of being buried. It is hypothesized herein that this ability is a result of the settling that occurs when a flowing avalanche consisting of snow and a large volume of air stops moving. On settling, most of the air would be forced out of the snow to the surface carrying the scent of a buried subject with it. The expired air would leave a trapped scent plume in the snow above the subject and scent on the surface of the snow that dogs could use to detect and locate them.
Frozen soils partially or completely block the movement of scent. When snow is present, probing the soil surface is the only sure way to determine if the soil is frozen or not. Contraction cracking caused by freezing or drying soils can allow scent to escape directly to the atmosphere provided the cracks penetrate to the buried source.
In addition to the difficulties noted above, others exist partly because of a lack of quantitative information on what constitutes optimal conditions and partly because of uncertainty in predicted weather conditions at the time of the proposed search. For the most part, we can only define search conditions qualitatively: those that have a lower probability of success and those that have a higher probability. We cannot state a reliable value for those probabilities.
The sources of VOCs evolving into the air above buried mines are the VOCs adsorbed on the surface soil particles (i.e. in the scent print), VOCs in the gas phase that transit the soil from below, and VOCs produced by volatilization at the soil surface of soil water. When scent reaches the ground surface, weather effects and other factors that influence volatilization at the ground surface and in ground covers determine the amount of scent available for detection by SDs (i.e. favorable search strategies, times, and conditions for searching for buried sources).
Favorable times to detect a buried source are when working and scenting conditions are good and when the scent is likely to be available to detect. Working and scenting conditions are good when the dogs do not become physically or mentally stressed and poor when they do. Examples of poor conditions include high temperatures, dusty and windy conditions, muddy ground, thick or thorny vegetation, steep terrain, as well as too much interference by the handler while the dog is searching. The availability of scent above the surface of a buried source depends on whether enough scent can survive passage through the soil and ground cover.
The size of the surface scent print is of interest to SD handlers since it helps to define the width of the search lanes to be used. Search lanes too wide reduce POD and search lanes too narrow expend more of the team’s energy and time than necessary. Lane widths for shallow buried sources appear to depend primarily on horizontal dimensions of the source and weather (especially atmospheric stability and wind, Figure 7). As the density of surface vegetation (grasses, weeds, prickly vegetation), depth of burial, and desired POD increases, lane widths should decrease. Wind makes it possible to use a greater lane width for a given set of conditions. The experience and reliability of the team is also important in selecting lane widths.
SDs used in demining operations are trained to work on long leads (about 9 to 12 yds) or on short leads. Long lead dogs work in lanes about 1.5 yds wide although the handler has some discretion in choosing the width. Short lead dogs work in lanes about 0.5 yds wide. The handlers always attempt to work the dogs in lanes across the wind and may change the lane direction with changes in wind direction. Other patterns that are used include circular, semi-random, and free ranging.
Limited information on lane widths for buried bodies and other large buried sources ranges from about 2 yds for poor conditions to about 10 yds or more for good conditions. Searches for bones, teeth, fired shell cases, small IEDs, and other small sources need smaller lane widths.
Linear searches (along paths, trails, roads, or natural features like vegetation, fields, shorelines, etc.) are best done on the downwind side of the feature being searched. A method often used in area searches by CD handlers is to first do a free ranging search through the area maximizing use of the wind and the team’s experience. Start the team on the downwind side, use a coarse grid, and check any likely places for a buried source. If nothing is detected, set up a fine grid with lanes running crosswind.
The properties of soils and VOCs and interactions between them indicate that care must be taken when extrapolating results from one site to another and from one type of source to another.
Table 2 compares the effects of soil conditions, source size, and depth of burial that result in lower or higher PODs for buried sources. Lower PODs do not mean that the source cannot be detected but that detection is less likely than higher PODs. Higher PODs occur when T, M, and P at the surface are less than at the source, in sand, or in gravel soils that are moist, with low organic contents, unfrozen, for large sources, and shallow burial.
Weather conditions that influence scent movement from buried sources to the atmosphere include wind, barometric pressure, air temperature, radiation, humidity, and precipitation, including snow (Table 3). These conditions are important because they influence scent transport processes through their effects on soil conditions at depth, at the ground surface, in ground cover, and in the air above. Higher PODs occur when winds are light (5 to 10 mph); barometer is falling; tide is rising; 32°F < T < 65°F; it is cloudy or sunny with light wind; humidity is >20%; there is dew or light rain; snow is dry powder, has never melted, has no wind slab, or is <1 f deep; it is during summer in late evening, night, or early morning, or fall, winter, or early spring.
Table 4 compares the effects of vegetation on PODs for buried sources. Higher PODs occur when there is no vegetation; when it is sparse, lower than the dog’s head; and when there is no prickly vegetation. Detection success in searches by EDs decreased with increasing vegetative cover near a mine. If the vegetation allowed the surface to dry and was sufficiently sparse so that the dogs could get their noses close to the ground, then a mine search could still be done.
Tom Osterkamp
0 comments:
Post a Comment